Abstract:
Traditional and new mainstream semiconductors are all sp3-coordinated crystals, including III-V, IV-IV, and, to a lesser extent, II-VI compounds, which are collectively referred to as octet compounds. Among octet compounds that exist in sp3-coordinated polymorphs, only boron nitride is known to exist also in sp2-coordinated forms, the most common of which is hexagonal BN (h-BN). Given the tremendous interest in two-dimensional (2D) crystals, a natural question is whether h-BN-like polymorphs can exist for at least some octet compounds other than BN. A theoretical study [1], based on an energetic consideration, predicted that each cation-anion bilayer in a wurtzite {0001} film would collapse into a planar,  h-BN-like structure if and only if the film thickness is below a certain threshold. This transformation to the nonpolar structure is deemed a new stabilization mechanism for the otherwise polar crystals to avert the polar field in the ultra-thin limit [1]; a multitude of known mechanisms counter the would-be catastrophic divergence of potential due to the polar field for bulk crystals [2,3]. While the h-BN-like ultra-thin films were hailed as “precursors to wurtzite films,” experimental evidences have been illusive despite efforts to grow ultra-thin films [4,5]. This talk presents the discovery of h-BeO, the h-BN-like form of BeO, made in a serendipitous experiment at CNM. Nanocrystals of BeO formed in graphene-sealed liquid cells were identified by HRTEM and EELS. Since h-BeO and the usual wurtzite BeO (w-BeO) have nearly identical basal plane lattice constants, we resorted to the “fine structure” of EELS, or energy loss near edge structure (ELNES), to show the sp2 electron configuration. Furthermore, we measured h-BeO thicknesses significantly larger than the thermodynamic threshold above which w-BeO is more stable. I will explain why this can be achieved, as well as why previous attempts did not lead to h-BN-like films. Our theoretical work further shows that the h-BN-like thin films of the octet compounds with wurtzite bulks are not so much like h-BN. They constitute a new type of 2D materials, of which we just had a first glimpse.

[1] C. L. Freeman et al, Graphitic nanofilms as precursors to wurtzite films: theory, Phys. Rev. Lett. 96, 066102 (2006).
[2] C. Noguera, Polar oxide surfaces, J. Phys. Condens. Matter 12, R367 (2000).
[3] A. Wander et al, Stability of polar oxide surfaces, Phys. Rev. Lett. 86, 3811 (2001).
[4] Tusche et al, Observation of depolarized ZnO(0001) monolayers: formation of unreconstructed planar sheets, Phys. Rev. Lett. 99, 026102 (2007).
[5] Lee et al, Tunable lattice constant and band gap of single- and few-layer ZnO, J. Phys Chem. Lett. 7, 1335 (2016).